Bioinformatic Tools in Crop Improvement - Crop Improvement: New Approaches and Modern Techniques - page 65

families by methods such as Markov Clustering (MCL), multi dimensional scaling
(MDS) and principle component analysis (PPO) (Table 2.1 ). Advanced statistics
can yield data sets that are useful for further variation analysis using gene mark-
ers, as well as phylogenetic studies, using Analysis of Co-Variance and Phylogeny
programs (see also Sect. 2.5). Listed in Table 2.1 are a number of web-based sites
for genetic variation analysis and computation; and for ease their purpose and URL
are also detailed there.
TranscriptomeResources
Comprehensive, high-throughput analysis of gene expression, also called 'transcrip-
tome' analysis, is a good approach to screen targeted genes, predict gene function
and discover cis -regulatory motifs. Hybridization-based methods, such as that used
in microarrays and GeneChips have been well established now for acquiring large-
scale gene expression profiles from various species. The recent rapid accumulation
of data containing large-scale gene expression profiles, and comparison of this data
to large repositories in genetic databanks have provided large amounts of informa-
tion now available in the public domain. This public data is an efficient and valuable
resource for many secondary uses, such as co-expression of genes and comparative
genomic studies. Furthermore, as next-generation DNA sequencing applications
and deep sequencing of short fragments of expressed RNAs and sRNAs become
common, they become important tools to use in both genome-sequenced and non-
sequenced species (Harbers and Carninci 2005 ; de Hoon and Hayashizaki 2008 ).
Listed below are the most important web-based sites for microchip and microarray
analysis; their purposes and their URL are detailed in Table 2.4 .
Sequence Tag Based Transcriptomics
Documentation of large-scale sequence ESTs from cDNA libraries was an early
approach in developing transcriptome data. The alternative is to use ESTs that are
randomly sequenced in an unbiased cDNA library, which are classified into clusters
of transcriptional units using sequence-clustering and/or other assembly methods.
The abundance of each transcript unit expressed in each tissue is then estimated by
counting the number of ESTs with identifiers for each cDNA library and/or each
sequence cluster. The same methodological principles have been applied in hu-
man and mouse, and a form of 'organism map' to determine the transcriptome in
various tissues and organs has been realised (Hishiki et al. 2000 ; Kawamoto et al.
2000 ; Ogasawara et al. 2006 ). There are no impediments in similar methods and
approaches being use in plant and crop transcriptomics.